Lund University's profile area for natural and artificial cognition welcomes you to its fifth fika-fika event about research on how aspects of human, animal, or artificial cognition, through communication and other forms of interaction, influence behaviour in the individual, dyadic, or collective context.

When: 19 October at 9.30 to 15.30 CET

Where: Centre for Mathematical Sciences, Lecture Hall Riesz, Märkesbacken 4, Lund, Sweden

Registration: ai.lu.se/2026-10-19/registration

10.00 Opening session

Introduction: Kalle Åstöm

Invited talk: Neurocognitive processing of human speech in dogs

Speaker: Lilla Magyari, Associate Professor, Psychology, University of Stavanger

10.50 Research presentation (NAC, Lund University)

Toward Generalist Open-World Temporal Perception

Speaker: Cristian Sminchisescu, Professor, Computer Vision and Machine Learning, Lund University

11.10 Excelence clusters - panel

Moderator: Melvyn B Davies, Mathematics, Lund University

• Self Evolving Software
  Per Runeson, Professor, Software Engineering, Lund University
• Robotics Institute Sweden
  Volker Krueger, Professor, Robotics and Semantic Systems, Lund University
• Connected Intelligence and Perception
• Fredrik Tufvesson, Professor, Communications Engineering, Lund University

12.00 Lunch

13.00 Afternoon session

Moderator: Marianne Gullberg, Professor, Linguistics, Lund University

Invited talk: Challenges in Fine-Grained Image Analysis

Speaker: Serge Belongie, Professor of Computer Science at the University of Copenhagen [abstract]

13.50 Research presentations (NAC, Lund University)

Irony Re-examined in Interactive Settings (IRIS): Towards a paradigm change in Experimental Pragmatics

Speaker: Camilo Rodriguez Ronderos, Assistant Professor, General Linguistics, Lund University [abstract]
 

Why K-9 students are better off in schools that are restrictive with students' use of GenAI in school

Speaker: Agneta Gulz, Professor, Cognitive Science [abstract]

Dynamics of Polarisation and Agreement in Socio-Technical Networks

Speakers: 

• Emma Tegling, Senior lecturer, Automatic Control
• Hanna Bäck, Political Science, Lund University

15.00 Fika & mingle

While waiting, you are welcome to check out recordings from last year's workshop, Natural and Artificial Cognition IV - 7 October 2025

Organisation and contact

• Hanna Bäck, Political Science
• Melvyn B Davies, Mathematics
• Marianne Gullberg, Linguistics
• Kalle Åström, Computer Vision and Machine Learning
• Jonas Wisbrant, AI Lund, NAC

Serge Belongie - abstract: Fine-grained image analysis (FGIA) is a longstanding and fundamental problem in computer vision and pattern recognition, underpinning a diverse set of real-world applications. The task of FGIA concerns visual objects from subordinate categories, e.g., species of birds or models of cars. Capitalizing on advances in Large Vision-Language Models (LVLMs), in recent years we have witnessed remarkable progress on a variety of FGIA problems. In this talk we review representative examples in the context of recognition and retrieval alongside other key issues such as publicly available benchmark datasets, domain-specific applications, and connections with other modalities including text and audio. We conclude by highlighting several research directions and open problems.

Camilo Rodriguez Ronderos - abstract: Understanding how language is used often involves going beyond a literal meaning by thinking about the intentions of speakers. While theoretical work in pragmatics typically aims to account for how addressees (intended recipients of an utterance) understand a speaker’s intentions, most experiments testing pragmatic theories have studied the comprehension of participants in laboratory experiments who do not take part in the conversation and are not the explicitly intended recipients of an utterance – the overhearers. A prime example of this is irony, a language strategy that relies on a contrast between the literal and intended meaning of an utterance (e.g., hearing someone say that a football player who missed an easy goal ‘took a beautiful shot’). The focus on the overhearer’s perspective has been particularly problematic for the study of irony, where experiments have resulted in contradictory and inconclusive results. 

Agneta Gulz - abstract: The talk will bring up some or all of:

• The difference between cognitive off-loading and cognitive out-sourcing
• The reasons why using GenAI profitably requires using it for things one already knows how to accomplish – in other words, not for learning things
• A large-scale study, presented in a preprint in June 26, titled “The Generative AI Learning Penalty: Evidence from Chinese Secondary Education” (by D. Strömberg, V. Lei, and Y. Wu).
• New York's pausing of student-facing GenAI in K-8 public schools"